Ndivhuwo Makondo
Papers
6
Total Citations
83
H-Index
5
About
Ndivhuwo Makondo is a leading researcher in robot learning, specializing in knowledge transfer, model learning, and imitation learning for robotic systems. His work addresses critical challenges in enabling robots—from manipulators to humanoids—to learn efficiently from limited data and to acquire skills from non-expert human demonstrations. Makondo’s major contributions include developing the Local Procrustes Analysis method for knowledge transfer across robots, which accelerates model learning by leveraging inter-robot data, as demonstrated in his highly cited 2015 paper (26 citations). He has also pioneered manifold mapping frameworks that allow robots with unknown kinematics to imitate human demonstrations in real-time, a breakthrough for diverse, real-world robotic applications. His research on model-based deep reinforcement learning further advances robot behavior acquisition by reducing the data burden of trial-and-error learning. With over 80 total citations across his top papers, Makondo’s work has significantly impacted the fields of robot learning and control, making him a key figure in the development of data-efficient, transferable robotic intelligence.
Research Focus
Key Achievements
Top Papers
- 1Knowledge transfer for learning robot models via Local Procrustes Analysis26 citations · 2015
- 2Modeling and control of a robot manipulator24 citations · 2013
- 3Accelerating Model Learning with Inter-Robot Knowledge Transfer14 citations · 2018
- 4Trajectory learning from human demonstrations via manifold mapping8 citations · 2016
- 5
- 6Knowledge Transfer using Model-Based Deep Reinforcement Learning3 citations · 2021